Communication method of road machine communication system of railway attended crossing

Abstract

The invention discloses a communication method of a road machine communication system of a railway attended crossing. According to the method, an Internet of Things camera is erected at a manned crossing; an early warning server and an intelligent reasoning server are erected in a machine room; a vehicle-mounted video terminal is erected on the train; a crossing handheld station device is configured for a worker at the attended crossing, and a complete system capable of guaranteeing the safety condition of the attended crossing of the railway is formed. According to the system, a 4G / 5G modern communication technology is utilized, machine vision and artificial intelligence technologies are fused, a crossing handheld station of a guarding person is connected with a train-mounted flat plate erected on a train, laser radars or cameras do not need to be frequently arranged along the railway to monitor the locomotive condition, and the locomotive condition can be monitored through communication between crossing handheld station equipment and the train-mounted flat plate. And the early warning server performs information transfer lane machine communication to complete safety protection work.

Description

Technical Field

[0001] The invention relates to a railway road machine communication system, and in particular to a communication method of a road machine communication system at a manned railway crossing. Background Art

[0002] my country has a vast territory and varied terrain. Railways not only run through mountainous and plateau areas, but sometimes inevitably cross urban or village crossings. For manned crossings that cross urban areas, when railway trains are running, people's vision is limited. Usually, drivers can only observe the safety of the crossing ahead through their own vision; when local crossing guards have not been confirmed by protection or an emergency occurs at the crossing, it is too late for the driver to take evasive measures after discovering the dangerous situation. This will greatly affect the safety of train driving and cause unnecessary loss of life and property to drivers, pedestrians, and railway companies.

[0003] At present, in rail transit at home and abroad, cameras or laser radars are used to implement safety monitoring at key locations such as crossings and stations. Frederic Maire of Queensland University of Technology abroad designed and simulated an image-based collision avoidance system model using Matlab based on the operation of Australian railway maintenance train formations. The on-board camera is used to obtain the view of the line in front of the train to estimate the distance between the train and the train in front, and prevent collision accidents between the two maintenance vehicles. However, the shooting distance of the on-board camera is limited, and its effective protection distance is within 140 meters. Such a distance is not enough for normal operating trains, especially for high-speed trains. Taking the CRH3 high-speed rail as an example, its common braking distance is more than 6,500 meters, and the emergency braking distance is also 3,500 meters. Therefore, it is necessary to provide the driver with images of several kilometers ahead to achieve the purpose of long-distance observation.

[0004] In China, there is a remote observation system for rail transit that uses original image sequences or image stitching and synthesis technology to realize the remote observation system of rail transit by arranging multiple cameras along the railway line to obtain the scenes within their respective monitoring ranges. This method sets up cameras along the locomotive route, integrating machine vision technology, image stitching and fusion technology, and modern communication technology to ensure the driving safety of rail transit. Although this long-distance observation technology does not rely on any speed measurement and positioning equipment, it is impossible to achieve intersection communication between the crossing and the locomotive in places such as downtown areas or village roads, lacking the protection of personal safety and people's property, and seriously affecting the safety of locomotives passing through manned crossings. Summary of the invention

[0005] In view of the problems and shortcomings of the above-mentioned prior art, the present invention provides a communication method for a railway manned crossing machine communication system.

[0006] The technical solution adopted by the present invention is: a communication method of a railway manned crossing machine communication system, the road machine communication system includes an early warning server, an intelligent reasoning server, a crossing handheld station, an Internet of Things camera, and a vehicle-mounted tablet; wherein the early warning server and the intelligent reasoning server are respectively connected to the crossing handheld station, the Internet of Things camera, and the vehicle-mounted tablet; The communication method of the road machine communication system has the following steps: 1. The early warning server receives the real-time position data of the train.

[0007] 2. The early warning server calculates whether the train has reached the preset alarm point in front of the crossing based on the received train position data.

[0008] 3. If it is determined that the preset alarm point in front of the crossing has been reached, the early warning server sends a protection confirmation message to the crossing handheld station, and the crossing handheld station provides feedback after receiving the protection confirmation message; if it is determined that the preset alarm point in front of the crossing has not been reached, return to step 2; the early warning server sends an alarm message to the vehicle-mounted tablet, and if no feedback information is received from the crossing handheld station device, it actively initiates a call to the crossing handheld station, and then executes step 6.

[0009] 4. The early warning server determines whether it has received the protection confirmation information feedback from the crossing handheld station. If it has, it executes step 6; if it has not, it transmits it to the vehicle-mounted tablet to actively call the crossing handheld station.

[0010] 5. The on-board tablet determines whether the call to the crossing handheld station is successful. If the call is successful, execute step 6. Otherwise, the train driver takes emergency protection to avoid an accident.

[0011] 6. The handheld station at the crossing completes the protection confirmation and uploads it to the early warning server. The early warning server determines whether the handheld station at the crossing has fed back the completion of the protection confirmation. If it has been completed, execute step 7; otherwise, return to step 4.

[0012] 7. The handheld station at the crossing sends the completed protection information to the vehicle-mounted tablet.

[0013] 8. The intelligent reasoning server pulls the video from the IoT camera at the front intersection and the video after intelligent reasoning of foreign objects, and sends it to the in-vehicle tablet display.

[0014] 9. The warning server returns to step 1 and executes steps 1 to 8 in a loop.

[0015] The beneficial effects of the present invention are as follows: the system sets up an Internet of Things (IOT) camera at a manned crossing; sets up an early warning server and an intelligent reasoning server in the machine room; sets up an on-board video terminal on the train; and configures a handheld device at the manned crossing to form a complete system that can ensure the safety of manned crossings on the railway. The system uses 4G / 5G modern communication technology, integrates machine vision and artificial intelligence technology, connects the handheld device of the crossing guard and the on-board tablet installed on the train, and does not need to frequently deploy laser radars or cameras along the railway to monitor the status of the locomotive. Through the communication between the handheld device at the crossing and the on-board tablet of the train, the early warning server performs information transfer and the communication between the road machine completes the safety protection work. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of the communication method of the road machine communication system of the present invention; Figure 2 It is a block diagram of the connection principle of the road machine communication system implementing the present invention. DETAILED DESCRIPTION

[0017] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0018] Reference Figure 2 The road machine communication system includes an early warning server, an intelligent reasoning server, a crossing handheld station, an Internet of Things camera, and a vehicle-mounted tablet; the early warning server and the intelligent reasoning server are respectively connected to the crossing handheld station, the Internet of Things camera, and the vehicle-mounted tablet (vehicle-mounted video terminal); among them, the early warning server and the intelligent reasoning server are set up in the computer room; the staff at the manned crossing are equipped with a crossing handheld station; the Internet of Things (IOT) camera is set up at the manned crossing; and the vehicle-mounted tablet is set up in the driver's cab at the head of the train.

[0019] The early warning server and intelligent reasoning server are set up in the computer room, the handheld equipment and IoT camera are set up at the crossing, and the on-board tablet is set up at the train. According to the location of the equipment, the system is composed of the computer room, the crossing, and the train. Data interaction exists between all the equipment in the system.

[0020] The crossing handheld station and the vehicle-mounted tablet have independent numbers (the crossing handheld station and the vehicle-mounted tablet have independent handheld station phone numbers and vehicle-mounted tablet phone numbers respectively). When initiating communication, a point-to-point direct communication method is adopted.

[0021] Reference Figure 1 , a communication method for a railway manned crossing machine communication system, the steps are as follows: 1. The early warning server receives the real-time position data of the train.

[0022] 2. The early warning server calculates whether the train has reached the preset alarm point in front of the crossing based on the received train position data.

[0023] 3. If it is determined that the preset alarm point in front of the crossing has been reached, the early warning server sends a protection confirmation message to the crossing handheld station, and the crossing handheld station provides feedback after receiving the protection confirmation message; if it is determined that the preset alarm point in front of the crossing has not been reached, return to step 2; the early warning server sends an alarm message to the vehicle-mounted tablet, and if no feedback information is received from the crossing handheld station device, it actively initiates a call to the crossing handheld station, and then executes step 6.

[0024] 4. The early warning server determines whether it has received the protection confirmation information feedback from the crossing handheld station. If it has, it executes step 6; if it has not, it transmits it to the vehicle-mounted tablet to actively call the crossing handheld station.

[0025] 5. The on-board tablet determines whether the call to the crossing handheld station is successful. If the call is successful, execute step 6. Otherwise, the train driver takes emergency protection to avoid an accident.

[0026] 6. The handheld station at the crossing completes the protection confirmation and uploads it to the early warning server. The early warning server determines whether the handheld station at the crossing has fed back the completion of the protection confirmation. If it has been completed, execute step 7; otherwise, return to step 4.

[0027] 7. The handheld station at the crossing sends the completed protection information to the vehicle-mounted tablet.

[0028] 8. The intelligent reasoning server pulls the video from the IoT camera at the front intersection and the video after intelligent reasoning of foreign objects, and sends it to the in-vehicle tablet display.

[0029] 9. The warning server returns to step 1 and executes steps 1 to 8 in a loop.

[0030] The early warning server continuously calculates the distance between the train and the crossing ahead, determines whether the locomotive has reached the preset point on the map line, and determines whether to trigger the early warning and protection functions after the calculation is completed. If the alarm is confirmed to be triggered, the alarm information will be sent to the on-board tablet device on the train; when the train approaches the crossing, the on-board tablet receives the alarm information from the early warning server, and then sends a protection confirmation message to the crossing handheld station. If no protection confirmation response is received from the crossing handheld station, it will automatically initiate a call to the corresponding crossing handheld station, and the main call will end after three failed calls.

[0031] The handheld station at the crossing is responsible for receiving protection alarms when the train enters the station, and reminds the crossing guards to complete the protection confirmation work through sound, light and vibration. After the staff completes the protection, they press the confirmation protection button to complete the protection work; the intelligent reasoning server is responsible for pulling the video information from the crossing camera ahead, and after intelligent foreign object recognition, it is output to the on-board tablet for the driver to refer to the situation at the crossing ahead.

[0032] When the locomotive is running, the early warning server receives real-time location information uploaded by the locomotive, continuously calculates the distance between the train and the crossing ahead, and sends protection information to the crossing handheld device when it detects that the train is approaching a preset warning point on the line; the crossing handheld device uses the Android 13 operating system, and is responsible for receiving protection confirmation, early warning and alarm functions sent by the early warning server. When the crossing guard finds an emergency at the crossing, he can initiate an active alarm to notify the vehicle about to enter the crossing of the emergency. When the locomotive is running, the crossing guard finds an emergency and reports it to the early warning server through the crossing handheld device. The early warning server sends an early warning notification to the vehicles near the alarm point, informing the corresponding vehicles of the emergency information ahead. After the early warning server calculates that the train has arrived at the preset point of the line, it will send a notification to the on-board tablet and display the telephone information of the crossing ahead on the on-board tablet. The Internet of Things (IOT) camera continuously collects video information from the crossing, and after the locomotive enters the preset point, it outputs the video to the on-board video terminal and the artificial intelligence reasoning server through the 4G / 5G network. The intelligent reasoning server receives the video information from the Internet of Things (IOT) camera and uses the YoloV7 algorithm for foreign object recognition. The results can be manually pulled and viewed on the on-board tablet and the crossing handheld device.

[0033] The handheld station at the crossing is carried by the guards on duty. When the train approaches the crossing, the guards manually click to confirm that the protection is complete. They can click to call the approaching train or be called by the train. The early warning server receives the location information of the locomotive, is responsible for determining whether the locomotive has reached the preset alarm point, and is responsible for sending protection confirmation information to the handheld station at the crossing. The Internet of Things (IOT) camera is responsible for collecting crossing video information and transmitting the video through the 4G / 5G network. After the locomotive arrives at the preset point, the intelligent inference server receives the crossing video information and runs the AI ​​foreign object recognition program, and the recognition results are output to the on-board video terminal. The on-board video terminal is responsible for receiving protection confirmation, can view the information of the crossing that is about to arrive ahead, and can initiate a call to the handheld station at the crossing ahead to complete the communication between the crossing and the locomotive.

[0034] The Internet of Things (IOT) camera is set up with a two-way panoramic monitoring camera based on the road conditions at the intersection.

[0035] The vehicle-mounted video terminal provided by the present invention adopts a tablet with an Android 13 operating system, which can display the real-time position of the locomotive and the information of the crossing ahead, and has physical buttons for easy operation by the driver.

[0036] The early warning server provided by the present invention is constructed for individual use by users, ensuring the most effective control of data, security and service quality by users. The physical server equipment that has been set up can be placed within the firewall of the user's enterprise data center, and sensitive data is controlled by the user end, which is a dedicated resource for the user.

[0037] The early warning server uses Lenovo SR588 4210R server, which has 32G running memory, 480GSSD + 12T*3 raid5 integrated board, 550W dual power supply, and Ubuntu22.04 distribution system for development. Ubuntu22.04 system is widely used in servers, with long official maintenance time and high stability. Deploying project-related services on it ensures the security and reliability of the system.

[0038] The intelligent reasoning server uses a server with an artificial intelligence reasoning graphics card (the graphics card model is NVIDIA T400), which is responsible for processing the intelligent reasoning of foreign object recognition and outputs the results to the vehicle video terminal after processing. The intelligent reasoning server uses the yolov7 algorithm for AI foreign object recognition services and pulls the Internet of Things (IOT) camera video when needed, effectively ensuring the control of user traffic costs.

Claims

1. A communication method for a railway manned crossing machine communication system, characterized in that: The road machine communication system includes an early warning server, an intelligent reasoning server, a road crossing handheld station, an Internet of Things camera, and a vehicle-mounted tablet; wherein the early warning server and the intelligent reasoning server are connected to the road crossing handheld station, the Internet of Things camera, and the vehicle-mounted tablet respectively; The communication method of the road machine communication system has the following steps:

1. The early warning server receives the real-time position data of the train; 2. The warning server calculates whether the train has reached the preset alarm point in front of the crossing based on the received train position data; 3. If it is determined that the preset alarm point in front of the crossing has been reached, the early warning server sends a protection confirmation message to the crossing handheld station, and the crossing handheld station gives feedback after receiving the protection confirmation message; if it is determined that the preset alarm point in front of the crossing has not been reached, return to step 2; the early warning server sends an alarm message to the vehicle-mounted tablet, and if no feedback information is received from the crossing handheld station device, a call is actively initiated to the crossing handheld station, and then step 6 is executed; Fourth, the warning server determines whether the protection confirmation information feedback from the crossing handheld station is received. If it is received, step 6 is executed; if it is not received, it is transmitted to the vehicle-mounted tablet to actively call the crossing handheld station; 5. The onboard tablet determines whether the call to the crossing handheld station is successful. If the call is successful, execute step 6. Otherwise, the train driver takes emergency protection to avoid accidents.

6. The handheld station at the crossing completes the protection confirmation and uploads it to the early warning server. The early warning server determines whether the handheld station at the crossing has fed back the protection confirmation. If it has been completed, execute step 7; otherwise, return to step 4; 7. The handheld station at the crossing sends the protection completion information to the vehicle-mounted tablet; 8. The intelligent reasoning server pulls the video from the IoT camera at the front crossing and the video after intelligent reasoning of foreign objects, and sends it to the in-vehicle tablet display; 9. The warning server returns to step 1 and executes steps 1 to 8 in a loop.

2. The communication method of the railway manned crossing machine communication system according to claim 1, characterized in that: The crossing handheld station and the vehicle-mounted tablet have independent numbers, and when initiating communication, a point-to-point direct communication method is adopted.

Images